MPTCP - Linux Kernel MultiPath TCP project


http://www.multipath-tcp.org/


MultiPath TCP (MPTCP) is an effort towards enabling the simultaneous use of several IP-addresses/interfaces by a modification of TCP that presents a regular TCP interface to applications, while in fact spreading data across several subflows. Benefits of this include better resource utilization, better throughput and smoother reaction to failures. Slides - explaining MultiPath TCP - are available in.pdf and .pptx format. You can also have a look at our Google Techtalk about MPTCP, or the MultiPath TCP tutorial given by Olivier Bonaventure at the IETF87.

The IP Networking Lab is implementing MPTCP in the Linux Kernel and hosting it on this website for users, testers and developers.

For questions, feedback,... please contact us at the mptcp-dev Mailing-List

If you use the implementation in scientific papers, please indicate in the paper the version of the implementation that you use and cite it as :

C. Paasch,S. Barre, et al., Multipath TCP in the Linux Kernel, available from http://www.multipath-tcp.org

Stable Release

MultiPath TCP v0.90 is available on ourrelease page.

Commercial support for Multipath TCP

Sébastien Barré and Gregory Detal, two of the keycontributors to the Multipath TCP implementation in the Linux kernel,are creating a company that will develop innovative products based onthis technology.

If you are working on a commercial application on top of Multipath TCPand need commercial-grade support, new features or backport toprevious kernel releases, just contactGregory Detal and Sébastien Barré directly to discuss with them about the services that they can offer.

The fastest TCP connection with Multipath TCP

Breaking the record of the fastest TCP connection - have a look here how we can achieve 51.8 Gbit/second with Multipath TCP.

Who is using MultiPath TCP?

Check out on our map, who has visited our website with MPTCP enabled. Statistics since September 2012.

The following websites are using the Linux MPTCP implementation. If you are using MPTCP on a public website, let us know atmptcp-dev Mailing List.

http://multipath-tcp.org
http://amiusingmptcp.com
http://ixit.cz
http://technosrix.com
http://watchy.in
http://hsh.link/
http://apt.tetaneutral.net
http://www.hchs.de/
http://www.neurocode-ag.com/
http://www.ip-mobilphone.net/
http://www.usenet-replayer.com/

If you are using MultiPath TCP (personnally or for business), let us know and share your experience on themptcp-dev Mailing List.

Apple's iOS7 supports Multipath TCP for its traffic generated by Siri. Congratulations to Apple for pushing the first wide-spread deployment of MPTCP.

Citrix's Netscaler as well as theF5 BIG-IP also support Multipath TCP!

Linux Kernel MultiPath TCP in action

We did a little demo of MultiPath TCP used over Ethernet/WiFi/3G on our Linux Kernel implementation.

We start an ssh-session with X-redirection and launch xscreensaver demo on the distant MPTCP-capable server.

We then turn off Ethernet and WiFi and thanks to MultiPath TCP the ssh-session is able to handover the traffic to 3G without interrupting the user-experience.Without our MPTCP Linux Kernel the session would simply stop working and the user would need to restart the ssh-session.

In the video you see a trafficmonitor on the right, displaying the instantaneous amount of traffic going over each interface. Top is Ethernet, middle is WiFi and bottom is 3G.


  • 1
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
### 回答1: Action-net是一种用于动作识别的多路径激励模型。它通过同时考虑多个视角和多个时间尺度的信息,提高了动作识别的准确性。模型的核心是一个多路径卷积神经网络,它可以从不同的视角和时间尺度提取特征。此外,模型还使用了一种新的激励机制,可以增强不同路径之间的信息交互,从而进一步提高了准确性。 ### 回答2: Action-Net是一个用于动作识别的多路径激发(Multipath Excitation)模型。动作识别是计算机视觉领域的一个重要任务,旨在从视频中自动识别和分类不同的动作。Action-Net通过采用多路径激发方法,提高了动作识别的准确性和性能。 多路径激发是一种模型设计技术,通过在网络架构中引入多个并行路径,每个路径分别学习不同的特征表示进行动作识别。这样的设计可以捕获到不同级别、不同尺度和不同分辨率的特征信息,提供更加全面准确的动作表示。 Action-Net利用了深度神经网络的强大表达能力,将视频输入分为多个时间段或空间尺度,并在每个时间段或空间尺度上构建不同的网络路径。每个路径在不同的时间尺度或空间分辨率上学习不同的特征表示,这些特征表示可以捕捉到动作的关键信息。 通过引入多路径激发机制,Action-Net可以同时学习到描述全局结构和细节特征的表示。此外,多路径激发还可以减轻网络在学习过程中的过拟合问题,提高了模型的泛化能力。 在实验中,Action-Net在各种动作识别任务上取得了优秀的性能。与传统的单一路径模型相比,Action-Net的准确性有了显著提升。多路径激发不仅可以用于动作识别,还可以应用于其他计算机视觉任务,如目标检测和图像分割。 综上所述,Action-Net是一种基于多路径激发的动作识别模型,通过引入多个并行路径学习不同的特征表示,提高了动作识别的准确性和性能。它具有较强的表达能力和泛化能力,在计算机视觉任务中有着广泛的应用前景。 ### 回答3: "action-net: multipath excitation for action recognition"是一种用于行为识别的多路径激励模型。 行为识别是计算机视觉中的一项重要任务,它旨在通过分析视频或图像序列来识别和理解人类的动作或行为。为了更好地解决这个问题,研究人员提出了许多方法,其中一个方法就是使用深度神经网络。 "action-net: multipath excitation for action recognition"是一种基于深度神经网络的模型,它通过多路径激励的方式来增强行为识别的性能。多路径激励是指通过引入多个并行的路径,每个路径都具有不同的特征提取能力,以捕捉不同方面的行为特征。 在"action-net"模型中,每个路径都是一个独立的卷积神经网络,这些网络在不同尺度和层次上进行特征提取。每个网络都通过激励机制来增强其对特定行为特征的敏感性。这种激励机制可以使网络在学习时集中注意力于关键的特征,提高行为识别的准确性。 通过多个路径的组合,"action-net"模型能够充分利用不同尺度和层次的特征信息,增强行为识别的鲁棒性和泛化能力。此外,多路径激励还能够有效地缓解过拟合问题,提高模型的泛化能力。 总之,"action-net: multipath excitation for action recognition"是一种基于深度神经网络的行为识别模型,通过多路径激励的方式有效地提高了行为识别的性能。它能够捕捉不同尺度和层次的行为特征,提高模型的鲁棒性和泛化能力。

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值